Lead Astray: ACLU of Michigan Investigation Has Found Irregularities in Flint’s Water Tests
As its analysis of water samples from more than 275 homes is wrapping up, researchers at Virginia Tech have come to a distressing conclusion:
Because of dangerously high lead levels, the city’s water is not safe to drink or use for cooking unless it is run through a filter designed to capture the toxic heavy metal.
That conclusion -- reached after a group of residents collected water samples from 277 homes and sent them to VT for analysis -- is in stark contrast to the position held by city officials and the Michigan Department of Environmental Quality (MDEQ), which maintain that the water is safe.
MDEQ officials declined to comment when contacted by the ACLU of Michigan seeking comment for this story.
Brad Wurfel, the department’s spokesman, did respond to questions from a reporter at the Flint Journal. Attempting to discredit the VT study, this is some of what Wurfel had to say:
“The issue here isn’t Flint’s water source or water plants. It’s the high number of older homes with lead pipes and lead service connections…. Folks who have concerns should get a water specialist to take a look at their home and see what they need to do to achieve peace of mind, because lead and copper are home plumbing problems that no water source can eliminate entirely.”
The Virginia Tech team responded to MDEQs assertions.
“During our sampling events in Flint homes, we are finding very high lead in other homes with modern lead free plumbing, which again points to city owned lead pipes and corrosive water as the problem,” according to a posting on the Flint Water Project website.
“On the basis of these facts, we consider MDEQ’s position to be both unscientific and irresponsible, and we stand by our recommendations to Flint consumers, that they immediately reduce their exposure to high lead in Flint’s water by implementing protective measures when using tap water for drinking or cooking.”
Previously, in response to questions from a Michigan Radio reporter following up on the revelation that one Flint family had hazardous-waste levels of lead flowing into their home, Wurfel had advised that one house’s problems was merely an outlier, and that the people in Flint could “relax” when it came to concerns about lead I their drinking water.
The MDEQ, however, has a vested interest in defending the quality of Flint’s water. After receiving much cleaner Detroit water for decades, Darnell Earley, an emergency manager appointed by Gov. Rick Snyder, decided in a cost-saving measure to switch to the highly corrosive Flint River in April 2014 as the city’s main water supply until the K pipeline is built.
According to Edwards – a MacArthur grant recipient who uncovered a widespread problem of lead in Washington D.C.’s water in the early 200s -- MDEQ and the city did not consider it necessary to address the highly corrosive nature of the river water to distribution-system pipes that include lead and iron. This failure, he said, is wreaking havoc with Flint’s infrastructure and causing the very high lead levels.
If data collected by Edwards and his team at VT, in conjunction with Flint citizens and the ACLU of Michigan are correct, how is it that the city and state are able to assert that Flint is in compliance with U.S. Environmental Protection Agency requirements regarding lead and copper levels in the city’s drinking water?
Our investigation has found a number of problems with the testing that the city conducted and the state provided oversight. In combination, the measures helped assure that the city could be able to claim compliance with Federal Regulations.
Failure to follow federal testing procedures
The Federal compliance monitoring consists of two six-month sampling rounds. From June through December 2014 the city tested 100 homes. By law, the same 100 homes were supposed to have been tested again during the second six-month round, which began Jan. 1 and ended June 30.
According to the “90th percentile test” used to determine compliance, if more than 10 percent of those samples are above the federal action level of 15 parts per billion, the city “failed” to meet the lead action level, and is required to immediately alert the public to the problem and start taking steps to reduce lead in water. If at least 90 percent of the samples are under 15 ppb, the city can claim the water meets federal safety standards.
According to documents obtained by the ACLU through Freedom of Information Act requests, the second round of testing ended with only 71 samples collected by the city, instead of the 100 minimum required during the first round. In response, the MDEQ decided after the fact, that, because Flint’s population is less than 100,000 only 60 homes would have to be tested.
Of the 71 homes that were sampled in 2015, only 13 were from the list of 100 tested in 2014. According to Edwards, who reviewed the documents obtained by the ACLU and helped provide analysis of the data they contain, doing so violates the federal law dictating how sample sites are selected.
The intent of the 90th Percentile Lead and Copper Test is to identify and test homes considered high risk because they have either lead service lines or plumbing inside the house that contains lead.
Even if the decision to reduce the number of homes tested in the second round was legitimate, however, the city still should have been required to re-test all the homes sampled in the first round, Edwards said.
That didn’t come close to happening.
Only 13 of the original homes tested in the first round were re-tested this year.
(As with other questions regarding testing procedures, city officials did not respond to questions from the ACLU of Michigan regarding its testing procedures submitted in writing last week.)
It wasn’t just that the city didn’t follow EPA procedures in terms of testing the same homes sampled in the previous round.
Significantly, all the homes that were resampled in 2015 were homes that tested low for lead in 2014. None of the homes that had the highest lead levels, above the 15 ppb action level, were resampled. These actions clearly violate the law, Edwards said.
The home with the highest lead tested in 2014 belongs to Andrew Armstrong, a retired social worker. His first round test registered a startling 37 ppb, more than twice the EPA action level.
It wasn’t just that the home didn’t get re-tested. The city was required by law to notify the Armstrong’s that lead in their water was well beyond the danger zone.
According to Armstrong, that didn’t happen.
He didn’t learn of the danger until he received a call from Edwards.
“The first we heard about it was when Marc called us,” said Armstrong in a phone interview. “He said ours was one of the worst levels the city had sampled.”
What does Edwards make of the fact that low-level homes were re-tested while those with high levels were not?
“This is called the ‘cherry picking” of sites to find low lead, and it is against the law,” he said.
It helps the city “miss” a lead in water problem throughout Flint, and claim compliance with the federal standards, but other tricks were used as well.
Zeroing in on Flushing Road
With only days to go before the time ran out on the second round of sampling, it wasn’t just a failure to have the requisite number of homes tested that posed a problem for flint officials.
Even though it wasn’t testing homes that had previously shown high lead levels, analysis of water samples by the state found that so many samples collected by the city were above the federal action level, it faced a very strong likelihood of failing to meet the action level, meaning officials would have to alert the public and take action to lower the lead in city water. Such actions could include encountering the massive expense of starting to replace the city’s lead service lines.
On June 25, five days before the deadline to complete the collection of second-round testing samples, the MDEQ’s Adam Rosenthal sent an email to Michael Glasgow, Flint's utilities administrator, alerting him to the problem.
“We hope you have 61 more lead/copper samples collected and sent to the lab by 6/30/15, and that they will be below the AL [Action Level] for lead,” wrote Rosenthal. “As of now with 39 results, Flint’s 90th percentile is over the AL for lead.”
What happened during those final days that allowed the city to claim that the final 30 homes tested under the allowable limit?
For one thing, it zeroed in on a single section of Flushing Ave. on Flint’s northwest side. Seven homes – nearly 25 percent of those tested at the end of June – were clustered together on Flushing. Residents interviewed by the ACLU of Michigan said they were approached and asked to have their water tested. One of the residents said that person was a neighbor who works for the city’s water department.
As with the other homes tested during that span, the Flushing Ave. homes came in well-below the action levels. None registered higher than 6 ppb, and three had no detectable levels of lead at all.
Why would the city focus on that particular street?
A clue came from one of the residents, Steve Saneterick, who said he recalled major work being done on the water main in front of his home within the past decade or so.
Among the documents obtained by the ACLU of Michigan through its FOIA requests was a map showing projects undertaken in recent years by the city. Close inspection of the map revealed that in 2007 the city replaced a long section of water main along Flushing Ave.
According to Edwards, that’s significant for two reasons. One is that, as explained in a recent peer-reviewed paper, water mains with iron corrosion also tend to have more lead in the water. The other is that it’s common practice to replace lead service lines when a water main is replaced.
“For both those reasons, you are likely to find much less lead in these homes,” explained Edwards. “If you wanted to minimize the likelihood of finding lead in water, this is where you would concentrate your sampling.”
Maps and Mysteries
As noted earlier, federal law requires that lead and copper testing be focused on the homes most likely to be a problem. Specifically, at least 50 percent of the homes must have a lead services line, which connects to a water main. In addition, 50 percent must of lead plumping or fixtures inside the home.
In its reports submitted to the state, Flint officials claimed that every single home tested in both rounds one and two had lead service lines.
However, when we requested that the city provide us with the documents that enabled it to make that determination, we received nothing that showed what material the services lines are made of.
Which means one of two things: Either the city has the documentation and is not making it public, or the documents don’t exist and it really can’t say with any certainty which homes have lead service lines and which don’t.
So, how the city is able to claim that every single home it’s tested has a lead service is, at this point, a mystery.
But it is not the only one.
The Walters home has been tested at least four times. Initial tests by the city found lead levels of 104 parts per billion. Another test – uncovered in documents obtained as part of this investigation -- found levels of 700 ppb. Walters, who keeps meticulous records of all her interactions with the city, said she was never made aware of that test. That would constitute another violation of the law.
The first test of her water’s home – the one that found lead levels of 104 ppb -- was supposed to have been included in the samples used for the EPA compliance testing.
However, that test was deleted from the sample pool, ostensibly because her home has a whole house water filter, according to a hand-written not on a document we obtained through FOIA.
But Walters said that, under instruction from the city, the filter was disabled before the test was conducted and should have been included.
It’s deletion from the sampling pool is yet another action that helped the city and state claim that Flint is in compliance with the federal Lead and Copper (LCR) rule.
Having lost her trust in state and local officials long before that came to light, Walters began searching elsewhere for help, eventually finding a bureaucrat she believes “actually cares”: Miguel Del Toral, a water expert with the EPA’s Region V office headquartered in Chicago.
After coming to Flint earlier this year to personally inspect the Walters’ home and make sure there was no source of lead in its internal plumbing, Del Toral helped connect Walters with Edwards at Virginia Tech. Together, they conducted an independent test that found lead levels as high as 13,500 pbb, more than two-and-a-half times the 5,000 pbb levels of lead at which water is classified as hazardous waste.
Walters also discovered that one of her children was lead poisoned. Just four, his IQ has been reduced, and he faces the prospect of learning disabilities and behavioral issues as a result of the exposure to lead.
It is true, as the MDEQ’s Wurfel said, that her home is an outlier of sorts – none of the other testing in Flint has found lead levels anywhere close to the maximum found in her family’s home.
On the other hand, it was her connection with Del Toral that led to him writing an internal memo that alleged numerous problems with the way Flint was treating its water and conducting its testing.
The contents of that memo were first reported by the ACLU of Michigan in July, and helped spur Edwards to obtain an emergency grant from the National Science Foundation to conduct the study under way in Flint for the past month.
Specifically, he alleged the city’s method – which is approved as part of the state’s protocol – requires that water be run for about four minutes prior to letting it sit for at least 6 hours before it is sampled. This method, known as “pre-flushing,” is the equivalent to finding a “loophole” in the law and exploiting it, he wrote.
There was not pre-flushing in the Virginia Tech tests. As a part of the Flint Water Project, Edwards and his team were able to directly compare the results of 20 homes previously tested by the city with the university’s method.
“We found that there was a statistically significant difference between the two methods,” said Edwards, with the city’s method resulting in generally reduced lead findings.
Edwards has said that “no legitimate scientist” would use the pre-flushing method, and has called for it to be banned.
As the result of its findings, the Virginia Tech team, contrary to assertions by the city and state that Flint’s water is safe to drink, is recommending that tap water only be used for cooking or drinking only if passes through a filter certified to remove lead. Absent such filtration, they recommend residents flush their waterlines for five minutes at a high flow rate every time before using.
“We do not issue this warning lightly, and note that our concern is based on several lines of evidence,” the Virginia Tech team posted on the Flint Water Study web site.
There is another recommendation made that is also not done lightly.
Given the fact that the city is supposed to have targeted only high risk homes, the VT researches say they are confounded by the fact that their testing has revealed a situation much worse than the city and state claim.
The researchers are demanding answers as to why that is.
“In our experience, following the EPA site selection criteria targeting homes with the highest risk for lead, the MDEQ sampling should have found much worse results than our sampling. Instead, MDEQ is asserting that the lead levels in Flint are much lower. Hence, we call on the U.S. EPA and others, to conduct a detailed audit of the 2014 and 2015 LCR sampling round overseen by MDEQ in Flint, to determine if it was conducted consistent with requirements of the law.”
We believe that what we’ve uncovered in our investigation so far only adds impetus to that demand.